Skip to main content
Mallory
← Back to News
ai-platform-securityinternet-facing-service-vulnerabilitywidely-deployed-product-advisory

GitLab Patches Critical AI Gateway Sandbox Escape Allowing Arbitrary Command Execution

Emerging Technology Risk · 17 sources · Updated 2mo ago · First seen Oct 2, 2026

GitLab Patches Critical AI Gateway Sandbox Escape Allowing Arbitrary Command Execution

GitLab released AI Gateway versions 19.2.4, 19.3.2, and 19.4.1 to fix CVE-2026-90970, a critical improper neutralization vulnerability rated 9.9 on CVSS v3.1. Under certain conditions, an authenticated user with Duo Agent Platform access could submit a specially crafted flow configuration, escape the prompt template sandbox, and execute arbitrary commands on the AI Gateway. GitLab conducted targeted customer outreach before announcing the patches; the supplied notices do not report exploitation.

GitLab strongly recommends that affected Self-Hosted AI Gateway installations upgrade immediately to the applicable patched version. GitLab-hosted AI Gateways have already been patched, so GitLab.com, GitLab Dedicated, and GitLab Self-Managed customers using those hosted gateways need no action. The Canadian Centre for Cyber Security reinforced the update guidance in advisory AV26-994, identifying affected releases preceding the three fixed versions and urging administrators to review GitLab’s guidance and apply necessary updates.

New surface, new adversariesTrack how attackers are adapting to this technology.
Track this risk
Event timeline

How this story unfolded

Events from the most recent confirmed update back to the earliest known activity.

9 events · Feb 2026 – Oct 4, 2026
  1. Reported Oct 5, 20262mo ago

    CSIRT Italia issues advisory on critical GitLab AI Gateway vulnerability

    CSIRT Italia/ACN issued an advisory warning that a critical GitLab AI Gateway vulnerability could allow an authenticated attacker to execute arbitrary code. It recommended updating vulnerable installations according to GitLab's security bulletin.

    Risolta vulnerabilità su GitLab AI Gateway - ACN
  2. Oct 4, 20262mo ago

    Guyana National CIRT issues GitLab AI Gateway advisory ADV2026_647

    Guyana National CIRT issued advisory ADV2026_647, identifying GitLab AI Gateway versions preceding 19.2.4, 19.3.2, and 19.4.1 in their respective release lines as affected. It recommended that users and administrators review GitLab's guidance and apply applicable updates.

    CIRT.GY | GitLab Security Advisory (October 4th, 2026)
  3. Oct 2, 20262mo ago

    CISA adds exploitation assessment for CVE-2026-90970

    CISA added an assessment to the CVE-2026-90970 record listing exploitation as "none," rather than public proof of concept or active exploitation. GitLab's advisory did not state whether the vulnerability had been exploited.

    GitLab Patches Critical 9.9 AI Gateway Flaw Allowing Command Execution on Self-Hosted Servers
  4. Oct 2, 20262mo ago

    Canadian Cyber Centre issues GitLab advisory AV26-994

    The Canadian Centre for Cyber Security issued advisory AV26-994 concerning the GitLab AI Gateway vulnerability and linked to GitLab's critical patch release. It encouraged users and administrators to review the guidance and apply necessary updates.

    GitLab security advisory (AV26-994) - Malware News - Malware Analysis, News and Indicators
  5. Reported Oct 2, 20262mo ago

    GitLab releases AI Gateway 19.2.4, 19.3.2, and 19.4.1 for critical flaw

    GitLab released AI Gateway versions 19.2.4, 19.3.2, and 19.4.1 to fix CVE-2026-90970, rated critical with a CVSS v3.1 score of 9.9, and urged affected Self-Hosted customers to upgrade immediately. Under certain conditions, an authenticated user with Duo Agent Platform access could use a crafted flow configuration to escape the prompt template sandbox and execute arbitrary commands on the gateway.

    GitLab AI Gateway Critical Patch Release: 19.2.4, 19.3.2, and 19.4.1 | GitLab Docs
  6. Reported Oct 2, 20262mo ago

    GitLab privately contacts Self-Hosted AI Gateway customers

    Before publishing its release announcement, GitLab conducted targeted outreach to Self-Hosted AI Gateway customers with upgrade guidance.

    GitLab AI Gateway Critical Patch Release: 19.2.4, 19.3.2, and 19.4.1 | GitLab Docs
  7. Reported Oct 2, 20262mo ago

    GitLab patches its hosted AI Gateways before public announcement

    GitLab deployed the security fix to GitLab-hosted AI Gateways before announcing the release. GitLab.com, GitLab Dedicated, and GitLab Self-Managed customers using those hosted gateways did not need to take action.

    GitLab AI Gateway Critical Patch Release: 19.2.4, 19.3.2, and 19.4.1 | GitLab Docs
  8. Reported Oct 2, 20262mo ago

    Researcher responsibly discloses CVE-2026-90970 to GitLab

    Researcher invisiblemeerkat responsibly disclosed CVE-2026-90970, an improper neutralization vulnerability in the GitLab AI Gateway custom flow prompt template. GitLab credited the researcher in its patch announcement.

    GitLab AI Gateway Critical Patch Release: 19.2.4, 19.3.2, and 19.4.1 | GitLab Docs
  9. Feb 2026

    GitLab patches earlier critical AI Gateway vulnerability CVE-2026-1868

    GitLab patched CVE-2026-1868, a CVSS 9.9 AI Gateway vulnerability that could let an authenticated user cause denial of service or execute code through a specially crafted flow definition. Like CVE-2026-90970, it was classified as a CWE-1336 template-engine weakness.

    GitLab ја поправа критичната ранливост 9.9 во AI Gateway, која овозможува извршување команди на самохостирани сервери | MKD-CIRT | Национален центар за одговор на компјутерски инциденти
Linked entities

Related entities

Vulnerabilities, threat actors, malware, products, organizations, and breaches Mallory has linked to this story.

Open in app →

Affected products

1 linked
Ai-Gateway

Organizations

1 linked
GitLab
Do you run this?Follow how adversaries are adapting to this technology, and where it touches your stack today.
Track This Risk
Mallory platform

See the full picture, correlated to your attack surface.

This page covers what’s public. Mallory adds the parts that aren’t: which of your assets are affected, which threat actors are using it right now, and what to deploy next.

Exposure mappingMap indicators from this story to your assets and identify affected systems in minutes.
Threat actor evidenceEvery observed campaign, victim, and pivot linked to actors named in this story.
Associated malwareMalware, exploits, and IOCs connected to the activity described here.
Detection signaturesYARA, Sigma, and Snort rules deployed to your SIEM as soon as they’re published.
Scheduled alertsGet matching new stories delivered to your team as they break, not the next morning.
AI threadsAsk questions about this story and take action on the answers.